zlacker

>>benhur+(OP)
This happens again and again. I have had that happen to my twitter account. I see this regulary on HN.

My suspicion is that this is mostly happening because platforms that big like google or twitter rely very heavily on machine learning and other AI related technology to ban people. Because honestly, the amount of spam and abuse that are likely happening on these platforms has to be mind boggling high.

So I get why they would try to automate bans.

But after years and years of regular high profile news of false positives, one would think they eventually would change something.

I mean the guy had direct business with Google going on....

Why would they continue like that. Isn't there one single PR person at Google?

>>p410n3+Y2
Well, frankly speaking, as an individual or a small company, you do not matter much, especially in comparison to the cost to get the problem fixed. While an organization grows larger, it has to employ lots of processes which are obviously not perfect to make things work. When it grows even larger, it has to make changes to existing processes, abolish some processes become no longer appropriate and introduce new processes over existing processes to serve their business better. Unavoidably more and more automation are introduced and eventually AI. All those changes seem to be really minor and clear and works in most of cases. Yep, I meant most cases, not all cases. Then suddenly, something really should work per everything standard and process stopped working and no one really knows why. So here comes the question, if you are the decision maker, your system works for 99.999999% maybe even 99.999999999% of your customers but not for those 1 maybe 10 customers, are you going to spend $$$$$$$$$$$$$$$$$$$$$ to get it fixed?